European Journal of Pharmaceutical Sciences 52 (2014) 12–20
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps
FA-loaded lipid drug delivery systems: Preparation, characterization and biological studies Claudia Carbone a,⇑, Agata Campisi b, Teresa Musumeci a, Giuseppina Raciti b, Roberta Bonfanti b, Giovanni Puglisi a a b
Laboratory of Drug Delivery Technology, Department of Drug Sciences, University of Catania, Viale A. Doria 6, 95125 Catania, Italy Section of Biochemistry, Department of Drug Sciences, University of Catania, Viale A. Doria 6, 95125 Catania, Italy
a r t i c l e
i n f o
Article history: Received 21 March 2013 Received in revised form 29 August 2013 Accepted 8 October 2013 Available online 26 October 2013 Keywords: Lipid nanoparticles DSC Ferulic acid Glioblastoma Cellular viability Immunocytochemistry
a b s t r a c t The main purpose of this research was to prepare and to characterize ferulic acid-loaded nanostructured lipid carrier (FA-NLC) to evaluate the cytotoxic effect on human glioblastoma cancer U87MG cells. First of all, the influence of different materials on mean size and homogeneity of NLC prepared by a low energy organic solvent-free method was investigated. Technological characterization (encapsulation efficiency, mean particle size, homogeneity and in vitro release profile) was performed on the selected NLC in comparison to others lipid carriers, nanoemulsion and SLN. Furthermore, the thermal behavior of NLC and SLN was investigated using Differential Scanning Calorimetry (DSC) in order to evaluate their structure. Biological studies (MTT bioassay and caspase-3 cleavage) on the selected NLC showed no cytotoxic effects of the unloaded tested NLC. Besides, the effectiveness of FA-loaded NLC was higher compared to the free drug. Cells treated with FA or FA-loaded NLC showed a greater effect compared to idebenone (IDE) or IDE-loaded NLC, respectively. These results strongly support that FA-loaded NLC could be potentially used for the treatment of glioblastoma. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction The ability of nanoscaled drug delivery systems (NDDS) to incorporate different drugs can be successfully used to optimize drug properties such as solubility, stability and safety. The increased efforts of worldwide researchers in the field of NDDS led to the use of a great variety of materials and strategies for the development of different nanoscaled systems. In particular, NDDS can be distinct in vesicular nanostructures (nanoemulsions,
Abbreviations: AFM, Atomic Force Microscopy; BBB, blood–brain barrier; CNS, Central Nervous System; CP, cetyl palmitate (cutina CP); DMSO, dimethyl sulfoxide; DSC, Differential Scanning Calorimetry; DT, difference between the melting and the onset temperatures; EDTA, ethylenediaminetetraacetic acid; EE%, percentage of encapsulation efficiency; FA, ferulic acid; FBS, fetal bovine serum; Gluta-MAX, supplement constituted of 200 mM L-alanyl-L-glutamine dipeptide in 0.85% NaCl; IDE, idebenone; IPM, isopropyl myristate; IPP, isopropyl palmitate; IPS, isopropyl stearate; MEM, Modified Eagle Medium; MTT assay, 3-[4,5-dimethylthiazol-2-yl]2,5-diphenyl tetrazolium bromide based colorimetric assay for cell growth; NDDS, nanoscaled drug delivery systems; NE, nanoemulsions; NLC, nanostructured lipid carriers; NLDDS, nanoscaled lipid drug delivery systems; PBS, phosphate buffered saline; PCS, Photon Correlation Spectroscopy; PDI, polidispersity index; PIT, phase inversion temperature; S.D., standard deviation; SLN, solid lipid nanoparticles; ZAve, mean particle size. ⇑ Corresponding author. Tel./fax: +39 095 738 4211. E-mail address: [email protected] (C. Carbone). 0928-0987/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ejps.2013.10.003
liposomes, niosomes) and nanoparticulate carriers, which include nanospheres and nanocapsules prepared with both polymers or lipids (Preetz et al., 2010). The use of liquid and solid lipids as matrices for the preparation of nanoscaled lipid drug delivery systems (NLDDS) guarantees many advantages in respect to other materials, in particular: good biocompatibility, low cytotoxicity, a good control in drug release, potential industrial scale-up and a wide range of applications (Müller et al., 2011). Different lipids can be employed for the production of lipid nanocarriers, such as triglycerides and their mixtures, fatty acids and/or waxes, which can led to different types of NLDDS such as nanoemulsions (NE), solid lipid nanoparticles (SLN) and nanostructured lipid carriers (NLC). NLDDS have been recently proposed for the systemic administration of lipophilic compound for the treatment of different diseases of the Central Nervous System (CNS) due to their extremely reduced mean size (Martins et al., 2012). Furthermore, recent studies showed that antioxidant and anti-inflammatory drugs (idebenone, ferulic acid, curcumin) are able to activate the apoptotic pathway in different type of cancers (Tai et al., 2011; Serafim et al., 2011; Karthikeyan et al., 2011; Kundu et al., 2012; Bandugula and Rajendra Prasad, 2013). Among different types of tumor, glioblastoma represents the most common and most aggressive form of brain tumor (Van Meir et al., 2010).
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
On the basis of these considerations, this work aimed to the preparation of different NLDDS (NE, SLN and NLC) to select the ideal carrier with the most adequate physico-chemical and technological properties for the delivery of ferulic acid (FA) for the treatment of glioblastoma. Photon Correlation Spectroscopy (PCS), Atomic Force Microscopy (AFM), Differential Scanning Calorimetry (DSC) and in vitro release studies have been performed for a better investigation of the physico-chemical and technological differences between the prepared carriers. All lipid nanocarriers have been prepared using a low energy organic solvent-free phase inversion method, that has received increasing interest due to the low energy required for the formation of the nanosystem compared to other techniques (ultrasound, high-pressure homogenization, dilution of microemulsion) (Anton et al., 2007). To achieve our goal we also investigated the effect of free FA and FA-loaded nanosystem selected among the prepared carriers, on the ability to induce cytotoxic effect on glioblastoma cancer cell lines (U87MG). Previously, a dose–response curve was developed to verify that the unloaded carrier did not affect cell cytotoxicity (Musumeci et al., 2006). In particular, we evaluated the cellular viability by MTT test and the activation of the apoptotic pathway by immunocytochemical analysis testing caspase-3 cleavage. In this study, idebenone (IDE) was also used for comparative purpose, both as free drug and loaded into the selected nanosystem, on the basis of literature data supporting its apoptotic activity and delivery into NDDS (Tai et al., 2011; Carbone et al., 2012a). 2. Materials and methods 2.1. Materials Cetyl palmitate (Cutina CP) was a gift from BASF Italia S.p.A. (Cesano Maderno, MB, Italy). Gliceryl Oleate (Tegin O), Oleth-20 (Brij 98) isopropyl myristate (IPM), isopropyl palmitate (IPP) and isopropyl stearate (IPS) were purchased from A.C.E.F. S.p.a. (Piacenza, Italy). Isoceteth-20 (Arlasolve 200), Sodium Pyruvate and TRITC-conjugated anti-mouse IgG polyclonal antibody, 3(4,5-dimethyl-thiazol-2-yl)2,5-diphenyl-tetrazolium bromide (MTT), TRITC-conjugated anti-mouse IgG polyclonal antibody, Lab-Tek II Chamber-Slide Systems, and others analytical chemicals were bought from Sigma–Aldrich (Milan, Italy). Ceteth-20 (Brij 58) and Ferulic acid (FA) were purchased from Fluka (Milan, Italy). Regenerated cellulose membranes (Spectra/Por CE; Mol. Wet. Cut off 3000) were supplied by Spectrum (Los Angeles, CA). Idebenone (IDE) was a kind gift of Wyeth Lederle (Catania, Italy). Methanol, acetic acid and water used in the HPLC procedures were of LC grade and were bought from Merck (Milan, Italy). All other reagents were of analytical grade. U87MG human glioblastoma cancer cell lines were purchased from Cell Bank Interlab Cell Line Collection (Genova, Italy). Modified Eagle Medium (MEM) with 2 mM GlutaMAX (GIBCO), Heath Inactivated Fetal Bovine Serum (FBS, GIBCO), Normal Goat Serum (NGS, GIBCO), non-essential amino acids, antibiotics, trypsin, Phosphate Buffer Saline solution (PBS) were from Invitrogen (Milano, Italy). Mouse monoclonal antibody against caspase-3 was from Becton Dickinson (Milan, Italy).
13
oleth-20). Three oils (IPM, IPP or IPS) were added to each formulation at 20% in respect to the total amount of solid and liquid lipid mixture (5% w/w) (Fig. 1). SLN and NE were obtained using the same surfactant mixture in combination with 8 or 5% w/w of wax or oil, respectively. The selected O3 formulation, then termed NLC-20%, was also studied with a lower amount (10%) of oil (NLC-10%) to better highlights thermodynamic differences between SLN and the NLC structure. To purify the nanoparticulate system from the excess of surfactants, each lipid nanoparticles sample was centrifuged at 10,000 rpm for 1 h (Beckman model J2-21 Centrifuge) and the nanoparticles re-dispersed in deionized water by vortex (60 s). 2.3. Phase inversion temperature determination The PIT temperature was determined using a conductivity meter mod. 525 (Crison, Modena, Italy), which measured electric conductivity change when the inversion from W/O to O/W system occurred. 2.4. Atomic Force Microscopy (AFM) The morphological analysis of the nanoparticulate systems was performed using a Park Autoprobe Atomic Force Microscope (Park Instruments, Sunnyvale, CA, USA). The analysis were performed in water at 20 °C and atmospheric pressure operating in non-contact mode (NC–AFM). Triangular silicon tips were used for this analysis. The resonant frequencies of this cantilever were found to be about 120 kHz. A drop of sample, diluted with water (about 1:50 v/v) was applied on a small mica disk (1 cm 1 cm); after two min the excess of water was removed and the sample was observed. Two images were obtained: the first one is a topographical and the second one is indicated as ‘‘error signal’’. This error signal is obtained by comparing two signals: the first, direct, representing the amplitude of the vibrations of the cantilever, and the other being the amplitude of a reference point. The images obtained by this method show small superficial variations of the samples. 2.5. Photon Correlation Spectroscopy (PCS) The particle size of all the prepared samples was determined by PCS which yields the mean particle size (ZAve) and the polidispersity index (PDI), that provides the width of the particles size distribution. PCS was performed with a Zetasizer Nano S90 (Malvern Instruments, Malvern, UK) at 25 °C, with a detection angle of 90° and a 4 mW He–Ne laser operating at 633 nm. Each sample was analyzed in triplicate. The results are shown as mean ± standard deviation (S.D.).
2.2. Preparation of NLDDS All NLDDS were prepared using the PIT method as reported in a previous work (Carbone et al., 2012b). NLC were prepared using three different primary surfactants: isoceteth-20 (10.6% w/w), ceteth-20 (8.7% w/w), oleth-20 (8.7% w/w) in combination with different percentage of the cosurfactant glyceryl oleate (3.5% w/w when combined with isoceteth-20; 4.4% w/w with ceteth-20 and
Fig. 1. Scheme of NLC formulations prepared in the preliminary screening of raw materials.
14
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
value was: r2 = 0.9998. No interference of the other formulation components was observed.
2.6. Differential Scanning Calorimetry (DSC) DSC scans of raw materials, freeze-dried empty and FA-loaded NLC and SLN (Freeze-dryer Edwards, Modulyo, 48 h) were performed on a Mettler DSC 12E equipped with a Haake thermocryostate model D8-G. A Mettler TA89E and FP89 system software was used for data acquisition. Indium was used to calibrate the instrument. The reference pan was filled with 40 or 100 lL of distilled water or empty pan. The raw materials were scanned in the temperature range 25–120 °C with a speed of 5 °C/min. The free drug, the freeze-dried empty and drug loaded samples were scanned at the same rate speed in the temperature range 25–225 °C. Each experiment was carried out in triplicate. 2.7. Encapsulation efficiency (EE%) The amount of the encapsulated drug into the lipid matrix of NLC and SLN has been determined after ultracentrifugation, dilution in methanol, vortex and filtration (0.22 lm). All samples have been analyzed by HPLC to evaluate the drug concentration. The encapsulation efficiency (EE%) has been calculated by the ratio between the quantity entrapped inside the nanoparticles and the total amount of drug used for their preparation (Eq. (1)):
EE% ¼ amount of drug entrapped=total amount of drug used 100 ð1Þ 2.8. In vitro release experiments FA release from NLC, SLN and NE containing FA 0.7% w/w was measured through regenerated cellulose membranes using Franz-type diffusion cells (LGA, Berkeley, CA) (Shah et al., 1989; Montenegro et al., 2011). The analysis was also performed on the NLC prepared with FA at 1.2% w/w. Before being mounted in Franz-type diffusion cells, the cellulose membranes were moistened by immersion in water for 1 h at room temperature. The receiving compartment (4.5 ml) was filled with a mixture 50:50 v/v of water/ethanol for ensuring pseudosink conditions by increasing active compound solubility in the receiving phase (Montenegro et al., 2011). As previously reported in literature, the presence of ethanol in the receiving phase did not compromise the nanoparticle integrity (Stancampiano et al., 2006). The receiving phase was constantly stirred at 700 rpm and thermostated at 37 °C. 500 ll of each sample were applied on the membrane surface (diffusion surface area of 0.75 cm2). The experiments were run for 24 h. At fixed time intervals, 200 ll of the receptor phase were withdrawn and replaced with an equal volume of receiving solution equilibrated to 37 °C. Each sample was analyzed by the HPLC method described below to determine the FA content. 2.9. HPLC analysis The HPLC analysis was performed at room temperature using a Hewlett–Packard model 1050 liquid chromatograph (Hewlett– Packard, Milan, Italy), equipped with a 20 ll Rheodyne model 7125 injection valve (Rheodyne, Cotati, CA) and an UV–VIS detector (Hewlett–Packard, Milan, Italy). A reversed-phase C18 column (Symmetry, 4.6 15 cm; Waters, Milan, Italy) was used for the analysis. A mixture of methanol/CH3COOH (5% v/v) (60:40 v/v) was used as mobile phase. The column effluent at a flow rate of 1 ml/min was monitored continuously at k = 320 nm. To construct a calibration curve, known amounts of FA (range 0.1–100 lg/ml) were dissolved in methanol and the absorption was determined for the standard solutions. The linear regression
2.10. In vitro biological tests 2.10.1. Cell culture U87MG cell lines were suspended, plated in flasks at a final density of 2 106 cells with MEM + 2 mM Gluta-MAX (Gibco) supplemented with 10% of fetal bovine serum (FBS), streptomycin (50 lg/ mL), penicillin (50 U/mL), 1% non-essential amino acids and 1 mM Sodium Pyruvate, and incubated at 37 °C in humidified atmosphere containing 5% CO2. The medium was replaced every 2 or 3 days. When the cultures were about 85–90% confluent, the cells were trypsinized by 0.05% trypsin and 0.53 mM EDTA solution and incubated for 5 min at 37 °C in humidified atmosphere containing 5% CO2. Trypsinization was stopped by adding 20% FBS, resuspended and plated in flasks fed with fresh basic complete media. Cells were seeded again at 1:4 density ratio and incubated at 37 °C in humidified atmosphere containing 5% CO2. 2.10.2. Treatment of the cultures A lot of U87MG cancer cell lines were placed at the final density of 60 104 cells/well of a 96-multiwell flat-bottomed 200-ll microplates and untreated or treated with different concentrations of free drugs, unloaded and drug-loaded NLC. We also treated some cultures with an equal volume of DMSO used for the solubilization of IDE and FA (35 lM). 2.10.3. MTT bioassay MTT bioassay was performed to monitor cell viability (Martins et al., 2012; Carbone et al., 2012b; Musumeci et al., 2006). A lot of cell lines cultures were exposed to free drugs, unloaded and FA-loaded (0.7% w/w) NLC-20% (20, 36, 40 lM) for 24, 48 or 72 h [4], in order to establish their optimal concentrations and the exposure times (Campisi et al., 2003). The study was accomplished by investigating cell death effects of the same tested concentrations of free IDE and IDE-loaded NLC, used for comparative purposes (Tai et al., 2011). At the end of treatment time, 20 ll of 0.5% of MTT salts in PBS were added to each well sample. After 1 h of incubation, the supernatant was removed and replaced with 100 ll dimethylsulfoxide and incubated at 37 °C in a humidified 5% CO2–95% air mixture for 1 h. The optical density of each well sample was measured with a microplate spectrophotometer reader (Titertek Multiskan; Flow Laboratories, Helsinki, Finland) at k = 570 nm. Three measurements for each samples were performed. Results were expressed as a percentage of the control (untreated cells), taken as 100%, to equalize the different values obtained. 2.10.4. Immunocytochemistry A lot of U87MG cancer cell lines were placed in Lab-Tek II Chamber-Slide Systems (Nalge Nunc International, Germany) at the final density of 0.5 105 and incubated at 37 °C in a humidified 5% CO2–95% air mixture. When the cell cultures arrived about at 80% of confluence, they were untreated or treated with the optimal concentration of free drugs (FA or IDE), unloaded and drug-loaded NLC and incubated at 37 °C in a humidified 5% CO2–95% air mixture for 24 h. After the treatment, the cell cultures were fixed for 20 min with 4% paraformaldehyde in 0.1 M PBS, they were treated with 1% NGS in PBS and incubated for 1 h at 37 °C in humidified air and 5% CO2, in order to block unspecific sites (Campisi et al., 2004). Then, they were washed three times with PBS, were successively added with mouse monoclonal antibody against caspase-3 (1:100) and incubated overnight at 37 °C in humidified air and 5% CO2. Cell cultures were then washed three times with PBS and incubated for 2 h with TRITC-conjugated anti-mouse IgG polyclonal antibody (1:64
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
in PBS). Finally, the cell cultures were washed three times with PBS, and the Lab-Tek II Chamber-Slide Systems were mounted in PBS/glycerol (50:50), and analyzed on a Leica fluorescent microscopy (Germany), in order to visualize caspase-3 positive cells. No non-specific staining of U87MG cancer cell lines was observed in control incubations in which the primary antibody was omitted (Campisi et al., 2003; Campisi et al., 2004; Campisi et al., 2012). 2.11. Data analysis Results of the technological characterization were expressed as the mean ± S.D. of three experiments and Student’s t-test was used to evaluate the significance of the difference between mean values. Biological data were statistically analyzed using one-way analysis of variance (one-way ANOVA) followed by post hoc Holm–Sidak test to estimate significant differences among groups. Data were reported as mean ± SD of four experiments in duplicate. In both statistical analysis, differences between groups were considered to be statistically significant for p < 0.05 and highly significant for p < 0.01. 3. Results and discussion A preliminary physico-chemical characterization was performed to identify the proper raw materials, among those selected on the basis of our previous studies on NE and SLN, for the preparation of NLC with small particle size and high homogeneity. 3.1. Preliminary screening on raw materials for NLC preparation To establish the feasibility to obtain NLC, a preliminary screening was performed on different surfactants and oils selected on the basis of previous experiments (Montenegro et al., 2006; Carbone et al., 2012b). As solid lipid matrix we chose CP due to its high safety for the systemic administration and its ability to obtain SLN (Lukowski et al., 2000; Yang et al., 2011). All NLC were prepared by a low energy organic solvent-free phase inversion process (PIT method). PIT values ranged from 65 to 73 °C for all the prepared NLC and were not significantly affected by the primary surfactant. Nevertheless, a slight increase of PIT temperature was observed with the increase of the oil lipophilicity (data not shown), thus suggesting that different interactions occurred between the liquid lipid and the other components. The effects of the selected variables were explored on both particle size and polidispersity by AFM and PCS analysis. As shown by the physico-chemical characterization (Fig. 2), the surfactant significantly affected the mean particle size, since all samples prepared with isoceteth-20 had a PDI > 0.3%, related with the presence of two or three peaks in size distribution. AFM images confirmed PCS results, revealing the presence of a heterogeneous system (Fig. 3A and B).
15
Both ceteth-20 and oleth-20 allowed the formation of NLC with a single peak in size distribution (PDI < 0.3%), even if oleth-20 generates particles with mean size smaller than those obtained with ceteth-20 (Fig. 2). A similar behavior was reported for SLN and was attributed to the presence of the unsaturated bond in both the chemical structure of surfactant and cosurfactant, which reduced the curvature radius at the interfaces favoring the formation of small particles (Montenegro et al., 2006). Student’s t-test was performed with a 99% level of confidence, to test whether the differences between ceteth-20 and oleth-20 were statistically significant. The difference between ZAve values were found to be highly significant (p < 0.01), thus encouraging the use of oleth-20, instead of ceteth-20, in the preparation of NLC. Instead of primary surfactants, it seems that the oil lipophilicity did not significantly affect particle size distribution, even if the use of IPS allowed to obtain the smallest particle size ( NE > SLN > NLC-20% (Fig. 8). It is possible that the addition of a little amount of oil could interact with the wax so as to reduce the path’s length that the drug must make for diffusion through the lipid matrix, thus determining greater amount of FA released after 24 h (Thatipamula et al., 2011). On the contrary, increasing the amount of oil content (NLC-20%), allowed to achieve a better control in drug release, as supported by DSC data. The release experiment performed on NLC loaded with FA 1.2% w/w, confirmed these data.
18
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
Fig. 7. Thermograms of FA-loaded freeze-dried SLN, NLC 10% and NLC 20%.
(Van Meir et al., 2010) (Table 2). Previously, unloaded NLC (with 20% of oil content) was studied at different concentrations (20, 36, 40 lM) to evaluate its effect on cell viability. The study was accomplished by investigating cell death effects of the same tested concentrations of free IDE and IDE-loaded NLC, used for comparative purposes (Tai et al., 2011). No significant difference between untreated and DMSO-treated culture cells was found. Unloaded NLC showed no cytotoxic effects on U87MG cell cultures at all the tested concentrations. Results of the observations through fluorescent microscope and MTT assay highlight significant changes in cell viability related to the tested concentrations and further differences between free drugs and drug-loaded NLC (Table 2). The optimal values of concentration and exposure time for the treatment were found to be respectively 36 lM and 24 h, which are in agreement with other findings (Tai et al., 2011; Karthikeyan et al., 2011). Data showed that at the selected concentration, FA was able to reduce cell viability much more than IDE, whose cytotoxic effect was previously reported on dopaminergic neuroblastoma cells (Tai et al., 2011). The increase of the cytotoxic effect observed when drugs were loaded
Table 2 MTT assay on U87MG human cell untreated or treated with DMSO or different concentrations (20, 36, 40 lM) of surfactant solution, free drugs (IDE and FA), unloaded and drug loaded NLC (containing 20% of oil in the lipid matrix). Results are expressed as a percentage of the control (untreated cells), taken as 100%, to equalize for different values between the treatments. Results are reported with standard deviation (±S.D.). Fig. 8. Release profile from FA-loaded NLDDS in comparison: NE, SLN and NLC with different oil content containing 0.7% w/w of FA (A); NLC with 10% and 20% of oil content containing FA 1.2% w/w (B).
Treatment
Control DMSO Surfactant solution IDE FA NLC IDE-NLC FA-NLC
Due to its small particle size and better control in drug release, NLC-20% was chosen for in vitro further investigation, in order to evaluate its efficacy on human glioblastoma cells (U87MG). 3.5. In vitro biological test To assess if free FA or FA-loaded NLC were able to induce cytotoxicity, MTT bioassay was performed on U87MG culture cell lines
* **
% Vitality cells ± S.D. 20 lM
36 lM
40 lM
100 99.97 ± 0.02** 99.95 ± 0.01** 91.19 ± 0.03* 99.94 ± 0.06* 99.07 ± 0.03* 89.96 ± 0.04* 79.43 ± 0.05*
100 99.76 ± 0.01** 99.98 ± 0.01** 68.89 ± 0.02* 69.81 ± 0.04** 99.04 ± 0.01** 51.25 ± 0.04** 38.96 ± 0.02**
100 99.81 ± 0.03** 99.91 ± 0.03** 87.89 ± 0.01* 84.58 ± 0.04** 71.04 ± 0.01** 72.86 ± 0.01** 64.64 ± 0.01*
Significant for p < 0.05. Highly significant for p < 0.01.
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
19
Fig. 9. A: Fluorescent microscopic analysis of caspase-3 cleavage in U87MG human cell lines in absence or in presence of DMSO (35 lM) or 36 lM of free IDE, free FA, unloaded NLC, IDE-loaded NLC or FA-loaded NLC, for 24 h; B: quantification and statistical analysis of caspase-3 positive U87MG human cell lines obtained and collected from 4 fields/coverslip of four separate experiments. Highly significant values for p < 0.01**.
into the selected NLC highlights the potentiality of the nanoparticulate system in increasing drug effectiveness. To verify if the drug alone or loaded into the NLC was able to activate the apoptotic pathway, we assessed caspase-3 cleavage by immunocytochemical analysis (Fig. 9A). No significant number of positive cells for caspase-3 in DMSO and unloaded NLC was observed (Fig. 9). In contrast, we observed a significant number of positive cells for caspase-3 in the cultures treated with FA or IDE as free drugs (Fig. 9A). The number of positive cells for caspase-3 significantly increased when the drugs were loaded into the NLC (Fig. 9), and resulted in a major induction of cell death in U87MG cancer cell lines. The effect of free FA or FA-loaded NLC appeared more evident when compared to free IDE or IDE-loaded NLC, respectively (Fig. 9). The quantification and statistical analysis of caspase-3 immunolabeling obtained and collected from 4 fields/coverslip of four separate experiments was reported in Fig. 9B. No non-specific staining of U87MG cell line cultures was observed in control incubations in which the primary antibody was omitted. Taken together these data demonstrate that FA is able to induce cytotoxic effects in human glioblastoma cancer cell lines activating the apoptotic pathway, and that the effects significantly increase in FA-loaded NLC. 4. Conclusion The low energy organic solvent-free phase inversion process used in this study allowed to obtain a NLC system with small particle size and high homogeneity when oleth-20 was used as
primary surfactant. DSC studies on SLN and NLC containing 10% or 20% of oil component in respect to the total lipid mixture, highlight the formation of a different lipid matrix related not only to the oil presence or not (NLC vs SLN) but also to the amount of the oil used for the preparation of the NLC. In vitro release studies showed a better control in FA release rate for NLC with the higher percentage of oil, which was comparable to those observed for SLN. MTT bioassay and caspase-3 cleavage performed on human glioblastoma cancer U87MG cells, showed no cytotoxic effect of the unloaded tested NLC and the activation of the apoptotic pathway caused by FA-loaded NLC, whose effectiveness was higher compared to the free drug. Furthermore, a greater effect was observed when cells were treated with FA or FA-loaded NLC, compared to IDE or IDE-loaded NLC. These results strongly support that FA-loaded NLC could be potentially used for the treatment of glioblastoma. Further studies are ongoing to better understand the molecular mechanism induced by free FA or FA-loaded-NLC. Acknowledgments The present work was financially supported by Lilly Foundation and Italian MIUR: PRIN 2010-2011 for University of Catania. We thanks BASF Italia S.p.A. for purchasing cetyl palmitate. References Anton, N. et al., 2007. Nano-emulsions and nanocapsules by the PIT method: an investigation on the role of the temperature cycling on the emulsion phase inversion. Int. J. Pharm. 344, 44–52.
20
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
Bandugula, V.R., Rajendra Prasad, N., 2013. 2-Deoxy-d-glucose and ferulic acid modulates radiation response signaling in non-small cell lung cancer cells. Tumor Biol. 34, 251–259. Campisi, A. et al., 2003. Glutamate-induced increases in transglutaminase activity in primary cultures of astroglial cells. Brain Res. 978, 24–30. Campisi, A. et al., 2004. Glutamate-evoked redox state alterations are involved in tissue transglutaminase upregulation in primary astrocyte cultures. FEBS Lett. 578, 80–84. Campisi, A. et al., 2012. Expression of tissue transglutaminase on primary olfactory ensheathing cells cultures exposed to stress conditions. Neurosci. Res. 72, 289– 295. Carbone, C. et al., 2012a. Chemical and technological delivery systems for idebenone: a review of literature production. Expert Opin. Drug Deliv. 9, 1377–1392. Carbone, C. et al., 2012b. Preparation and optimization of PIT solid lipid nanoparticles via statistical factorial design. Eur. J. Med. Chem. 49, 110–117. Karthikeyan, S. et al., 2011. Radiosensitizing effect of ferulic acid on human cervical carcinoma cells in vitro Toxicol. In Vitro 25, 1366–1375. Kaur, I.P. et al., 2008. Potential of solid lipid nanoparticles in brain targeting. J. Control. Release 127, 97–109. Kovacevic, S. et al., 2011. Polyhydroxysurfactants for the formulation of lipid nanoparticles (SLN and NLC): effects on size, physical stability and particle matrix structure. Int. J. Pharm. 406, 163–172. Kundu, P., Mohanty, C., Sahoo, S.K., 2012. Antiglioma activity of curcumin-loaded lipid nanoparticles and its anhanced bioavailability in brain tissue for effective glioblastoma therapy. Acta Biomat. 8, 2670–2687. Lukowski, G. et al., 2000. Crystallographic investigation of cetylpalmitate solid lipid nanoparticles. Int. J. Pharm. 196, 201–205. Martins, S. et al., 2012. Brain delivery of camptothecin by means of solid lipid nanoparticles: formulation design, in vitro and in vivo studies. Int. J. Pharm. 439, 49–62. Montenegro, L. et al., 2006. Effect of oil phase lipophilicity on in vitro drug release from O/W microemulsions with low surfactant content Drug Dev. Ind. Pharm. 32, 539–548.
Montenegro, L. et al., 2011. In vitro evaluation of idebenone-loaded solid lipid nanoparticles for drug delivery to the brain. Drug Dev. Ind. Pharm. 37, 737–746. Müller, R.H., Radtke, M., Wissing, S.A., 2002. Nanostructured lipid matrices for improved microencapsulation of drugs. Int. J. Pharm. 242, 121–128. Müller, R.H., Shegokar, R., Keck, C.M., 2011. 20 Years of lipid nanoparticles (SLN & NLC): present state of development and industrial application. Curr. Drug Discov. Technol. 8, 207–227. Musumeci, T., Vicari, L., Ventura, C.A., Gulisano, M., Pignatello, R., Puglisi, G., 2006. Lyoprotected nanosphere formulations for paclitaxel controlled delivery. J. Nanosci. Nanotechnol. 6, 3118–3125. Preetz, C. et al., 2010. Application of Atomic Force Microscopy and ultrasonic resonator technology on nanoscale: distinction of nanoemulsions from nanocapsules. Eur. J. Pharm. Sci. 39, 141–151. Serafim, T.L. et al., 2011. Oliveira, Lipophilic caffeic and ferulic acid derivatives presenting cytotoxicity against human breast cancer cells. Chem. Res. Toxicol. 24, 763–774. Ò Ò Severino, P. et al., 2012. Crystallinity of Dynasan 114 and Dynasan 118 matrices for the production of stable Miglyol-loaded nanoparticles. J. Therm. Anal. Calorim. 108, 101–108. Shah, V.P. et al., 1989. Determination on in vitro drug release from hydrocortisone creams. Int. J. Pharm. 53, 53–59. Stancampiano, A.H.S. et al., 2006. Technological and biological characterization of idebenone-loaded solid lipid nanoparticles prepared by a modified solvent injection technique. J. Biomed. Nanotech. 2, 53–260. Tai, K.K., Pham, L., Truong, D.D., 2011. Idebenone apoptotic cell death in the human dopaminergic neuroblastoma SHSY-5Y cells. Neurotox. Res. 20, 321–328. Thatipamula, R.P. et al., 2011. Formulation and in vitro characterization of domperidone loaded solid lipid nanoparticles and nanostructured lipid carriers. DARU 19, 23–32. Van Meir, E.G. et al., 2010. Exciting new advantages in neuro-oncology: the avenue to a cure for malignant glioma. CA Cancer J. Clin. 60, 166–193. Yang, R. et al., 2011. The influence of lipid characteristics on the formation, in vitro release, and in vivo absorption of protein-loaded SLN prepared by the double emulsion process. Drug Dev. Ind. Pharm. 37, 139–148.
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps
FA-loaded lipid drug delivery systems: Preparation, characterization and biological studies Claudia Carbone a,⇑, Agata Campisi b, Teresa Musumeci a, Giuseppina Raciti b, Roberta Bonfanti b, Giovanni Puglisi a a b
Laboratory of Drug Delivery Technology, Department of Drug Sciences, University of Catania, Viale A. Doria 6, 95125 Catania, Italy Section of Biochemistry, Department of Drug Sciences, University of Catania, Viale A. Doria 6, 95125 Catania, Italy
a r t i c l e
i n f o
Article history: Received 21 March 2013 Received in revised form 29 August 2013 Accepted 8 October 2013 Available online 26 October 2013 Keywords: Lipid nanoparticles DSC Ferulic acid Glioblastoma Cellular viability Immunocytochemistry
a b s t r a c t The main purpose of this research was to prepare and to characterize ferulic acid-loaded nanostructured lipid carrier (FA-NLC) to evaluate the cytotoxic effect on human glioblastoma cancer U87MG cells. First of all, the influence of different materials on mean size and homogeneity of NLC prepared by a low energy organic solvent-free method was investigated. Technological characterization (encapsulation efficiency, mean particle size, homogeneity and in vitro release profile) was performed on the selected NLC in comparison to others lipid carriers, nanoemulsion and SLN. Furthermore, the thermal behavior of NLC and SLN was investigated using Differential Scanning Calorimetry (DSC) in order to evaluate their structure. Biological studies (MTT bioassay and caspase-3 cleavage) on the selected NLC showed no cytotoxic effects of the unloaded tested NLC. Besides, the effectiveness of FA-loaded NLC was higher compared to the free drug. Cells treated with FA or FA-loaded NLC showed a greater effect compared to idebenone (IDE) or IDE-loaded NLC, respectively. These results strongly support that FA-loaded NLC could be potentially used for the treatment of glioblastoma. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction The ability of nanoscaled drug delivery systems (NDDS) to incorporate different drugs can be successfully used to optimize drug properties such as solubility, stability and safety. The increased efforts of worldwide researchers in the field of NDDS led to the use of a great variety of materials and strategies for the development of different nanoscaled systems. In particular, NDDS can be distinct in vesicular nanostructures (nanoemulsions,
Abbreviations: AFM, Atomic Force Microscopy; BBB, blood–brain barrier; CNS, Central Nervous System; CP, cetyl palmitate (cutina CP); DMSO, dimethyl sulfoxide; DSC, Differential Scanning Calorimetry; DT, difference between the melting and the onset temperatures; EDTA, ethylenediaminetetraacetic acid; EE%, percentage of encapsulation efficiency; FA, ferulic acid; FBS, fetal bovine serum; Gluta-MAX, supplement constituted of 200 mM L-alanyl-L-glutamine dipeptide in 0.85% NaCl; IDE, idebenone; IPM, isopropyl myristate; IPP, isopropyl palmitate; IPS, isopropyl stearate; MEM, Modified Eagle Medium; MTT assay, 3-[4,5-dimethylthiazol-2-yl]2,5-diphenyl tetrazolium bromide based colorimetric assay for cell growth; NDDS, nanoscaled drug delivery systems; NE, nanoemulsions; NLC, nanostructured lipid carriers; NLDDS, nanoscaled lipid drug delivery systems; PBS, phosphate buffered saline; PCS, Photon Correlation Spectroscopy; PDI, polidispersity index; PIT, phase inversion temperature; S.D., standard deviation; SLN, solid lipid nanoparticles; ZAve, mean particle size. ⇑ Corresponding author. Tel./fax: +39 095 738 4211. E-mail address: [email protected] (C. Carbone). 0928-0987/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ejps.2013.10.003
liposomes, niosomes) and nanoparticulate carriers, which include nanospheres and nanocapsules prepared with both polymers or lipids (Preetz et al., 2010). The use of liquid and solid lipids as matrices for the preparation of nanoscaled lipid drug delivery systems (NLDDS) guarantees many advantages in respect to other materials, in particular: good biocompatibility, low cytotoxicity, a good control in drug release, potential industrial scale-up and a wide range of applications (Müller et al., 2011). Different lipids can be employed for the production of lipid nanocarriers, such as triglycerides and their mixtures, fatty acids and/or waxes, which can led to different types of NLDDS such as nanoemulsions (NE), solid lipid nanoparticles (SLN) and nanostructured lipid carriers (NLC). NLDDS have been recently proposed for the systemic administration of lipophilic compound for the treatment of different diseases of the Central Nervous System (CNS) due to their extremely reduced mean size (Martins et al., 2012). Furthermore, recent studies showed that antioxidant and anti-inflammatory drugs (idebenone, ferulic acid, curcumin) are able to activate the apoptotic pathway in different type of cancers (Tai et al., 2011; Serafim et al., 2011; Karthikeyan et al., 2011; Kundu et al., 2012; Bandugula and Rajendra Prasad, 2013). Among different types of tumor, glioblastoma represents the most common and most aggressive form of brain tumor (Van Meir et al., 2010).
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
On the basis of these considerations, this work aimed to the preparation of different NLDDS (NE, SLN and NLC) to select the ideal carrier with the most adequate physico-chemical and technological properties for the delivery of ferulic acid (FA) for the treatment of glioblastoma. Photon Correlation Spectroscopy (PCS), Atomic Force Microscopy (AFM), Differential Scanning Calorimetry (DSC) and in vitro release studies have been performed for a better investigation of the physico-chemical and technological differences between the prepared carriers. All lipid nanocarriers have been prepared using a low energy organic solvent-free phase inversion method, that has received increasing interest due to the low energy required for the formation of the nanosystem compared to other techniques (ultrasound, high-pressure homogenization, dilution of microemulsion) (Anton et al., 2007). To achieve our goal we also investigated the effect of free FA and FA-loaded nanosystem selected among the prepared carriers, on the ability to induce cytotoxic effect on glioblastoma cancer cell lines (U87MG). Previously, a dose–response curve was developed to verify that the unloaded carrier did not affect cell cytotoxicity (Musumeci et al., 2006). In particular, we evaluated the cellular viability by MTT test and the activation of the apoptotic pathway by immunocytochemical analysis testing caspase-3 cleavage. In this study, idebenone (IDE) was also used for comparative purpose, both as free drug and loaded into the selected nanosystem, on the basis of literature data supporting its apoptotic activity and delivery into NDDS (Tai et al., 2011; Carbone et al., 2012a). 2. Materials and methods 2.1. Materials Cetyl palmitate (Cutina CP) was a gift from BASF Italia S.p.A. (Cesano Maderno, MB, Italy). Gliceryl Oleate (Tegin O), Oleth-20 (Brij 98) isopropyl myristate (IPM), isopropyl palmitate (IPP) and isopropyl stearate (IPS) were purchased from A.C.E.F. S.p.a. (Piacenza, Italy). Isoceteth-20 (Arlasolve 200), Sodium Pyruvate and TRITC-conjugated anti-mouse IgG polyclonal antibody, 3(4,5-dimethyl-thiazol-2-yl)2,5-diphenyl-tetrazolium bromide (MTT), TRITC-conjugated anti-mouse IgG polyclonal antibody, Lab-Tek II Chamber-Slide Systems, and others analytical chemicals were bought from Sigma–Aldrich (Milan, Italy). Ceteth-20 (Brij 58) and Ferulic acid (FA) were purchased from Fluka (Milan, Italy). Regenerated cellulose membranes (Spectra/Por CE; Mol. Wet. Cut off 3000) were supplied by Spectrum (Los Angeles, CA). Idebenone (IDE) was a kind gift of Wyeth Lederle (Catania, Italy). Methanol, acetic acid and water used in the HPLC procedures were of LC grade and were bought from Merck (Milan, Italy). All other reagents were of analytical grade. U87MG human glioblastoma cancer cell lines were purchased from Cell Bank Interlab Cell Line Collection (Genova, Italy). Modified Eagle Medium (MEM) with 2 mM GlutaMAX (GIBCO), Heath Inactivated Fetal Bovine Serum (FBS, GIBCO), Normal Goat Serum (NGS, GIBCO), non-essential amino acids, antibiotics, trypsin, Phosphate Buffer Saline solution (PBS) were from Invitrogen (Milano, Italy). Mouse monoclonal antibody against caspase-3 was from Becton Dickinson (Milan, Italy).
13
oleth-20). Three oils (IPM, IPP or IPS) were added to each formulation at 20% in respect to the total amount of solid and liquid lipid mixture (5% w/w) (Fig. 1). SLN and NE were obtained using the same surfactant mixture in combination with 8 or 5% w/w of wax or oil, respectively. The selected O3 formulation, then termed NLC-20%, was also studied with a lower amount (10%) of oil (NLC-10%) to better highlights thermodynamic differences between SLN and the NLC structure. To purify the nanoparticulate system from the excess of surfactants, each lipid nanoparticles sample was centrifuged at 10,000 rpm for 1 h (Beckman model J2-21 Centrifuge) and the nanoparticles re-dispersed in deionized water by vortex (60 s). 2.3. Phase inversion temperature determination The PIT temperature was determined using a conductivity meter mod. 525 (Crison, Modena, Italy), which measured electric conductivity change when the inversion from W/O to O/W system occurred. 2.4. Atomic Force Microscopy (AFM) The morphological analysis of the nanoparticulate systems was performed using a Park Autoprobe Atomic Force Microscope (Park Instruments, Sunnyvale, CA, USA). The analysis were performed in water at 20 °C and atmospheric pressure operating in non-contact mode (NC–AFM). Triangular silicon tips were used for this analysis. The resonant frequencies of this cantilever were found to be about 120 kHz. A drop of sample, diluted with water (about 1:50 v/v) was applied on a small mica disk (1 cm 1 cm); after two min the excess of water was removed and the sample was observed. Two images were obtained: the first one is a topographical and the second one is indicated as ‘‘error signal’’. This error signal is obtained by comparing two signals: the first, direct, representing the amplitude of the vibrations of the cantilever, and the other being the amplitude of a reference point. The images obtained by this method show small superficial variations of the samples. 2.5. Photon Correlation Spectroscopy (PCS) The particle size of all the prepared samples was determined by PCS which yields the mean particle size (ZAve) and the polidispersity index (PDI), that provides the width of the particles size distribution. PCS was performed with a Zetasizer Nano S90 (Malvern Instruments, Malvern, UK) at 25 °C, with a detection angle of 90° and a 4 mW He–Ne laser operating at 633 nm. Each sample was analyzed in triplicate. The results are shown as mean ± standard deviation (S.D.).
2.2. Preparation of NLDDS All NLDDS were prepared using the PIT method as reported in a previous work (Carbone et al., 2012b). NLC were prepared using three different primary surfactants: isoceteth-20 (10.6% w/w), ceteth-20 (8.7% w/w), oleth-20 (8.7% w/w) in combination with different percentage of the cosurfactant glyceryl oleate (3.5% w/w when combined with isoceteth-20; 4.4% w/w with ceteth-20 and
Fig. 1. Scheme of NLC formulations prepared in the preliminary screening of raw materials.
14
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
value was: r2 = 0.9998. No interference of the other formulation components was observed.
2.6. Differential Scanning Calorimetry (DSC) DSC scans of raw materials, freeze-dried empty and FA-loaded NLC and SLN (Freeze-dryer Edwards, Modulyo, 48 h) were performed on a Mettler DSC 12E equipped with a Haake thermocryostate model D8-G. A Mettler TA89E and FP89 system software was used for data acquisition. Indium was used to calibrate the instrument. The reference pan was filled with 40 or 100 lL of distilled water or empty pan. The raw materials were scanned in the temperature range 25–120 °C with a speed of 5 °C/min. The free drug, the freeze-dried empty and drug loaded samples were scanned at the same rate speed in the temperature range 25–225 °C. Each experiment was carried out in triplicate. 2.7. Encapsulation efficiency (EE%) The amount of the encapsulated drug into the lipid matrix of NLC and SLN has been determined after ultracentrifugation, dilution in methanol, vortex and filtration (0.22 lm). All samples have been analyzed by HPLC to evaluate the drug concentration. The encapsulation efficiency (EE%) has been calculated by the ratio between the quantity entrapped inside the nanoparticles and the total amount of drug used for their preparation (Eq. (1)):
EE% ¼ amount of drug entrapped=total amount of drug used 100 ð1Þ 2.8. In vitro release experiments FA release from NLC, SLN and NE containing FA 0.7% w/w was measured through regenerated cellulose membranes using Franz-type diffusion cells (LGA, Berkeley, CA) (Shah et al., 1989; Montenegro et al., 2011). The analysis was also performed on the NLC prepared with FA at 1.2% w/w. Before being mounted in Franz-type diffusion cells, the cellulose membranes were moistened by immersion in water for 1 h at room temperature. The receiving compartment (4.5 ml) was filled with a mixture 50:50 v/v of water/ethanol for ensuring pseudosink conditions by increasing active compound solubility in the receiving phase (Montenegro et al., 2011). As previously reported in literature, the presence of ethanol in the receiving phase did not compromise the nanoparticle integrity (Stancampiano et al., 2006). The receiving phase was constantly stirred at 700 rpm and thermostated at 37 °C. 500 ll of each sample were applied on the membrane surface (diffusion surface area of 0.75 cm2). The experiments were run for 24 h. At fixed time intervals, 200 ll of the receptor phase were withdrawn and replaced with an equal volume of receiving solution equilibrated to 37 °C. Each sample was analyzed by the HPLC method described below to determine the FA content. 2.9. HPLC analysis The HPLC analysis was performed at room temperature using a Hewlett–Packard model 1050 liquid chromatograph (Hewlett– Packard, Milan, Italy), equipped with a 20 ll Rheodyne model 7125 injection valve (Rheodyne, Cotati, CA) and an UV–VIS detector (Hewlett–Packard, Milan, Italy). A reversed-phase C18 column (Symmetry, 4.6 15 cm; Waters, Milan, Italy) was used for the analysis. A mixture of methanol/CH3COOH (5% v/v) (60:40 v/v) was used as mobile phase. The column effluent at a flow rate of 1 ml/min was monitored continuously at k = 320 nm. To construct a calibration curve, known amounts of FA (range 0.1–100 lg/ml) were dissolved in methanol and the absorption was determined for the standard solutions. The linear regression
2.10. In vitro biological tests 2.10.1. Cell culture U87MG cell lines were suspended, plated in flasks at a final density of 2 106 cells with MEM + 2 mM Gluta-MAX (Gibco) supplemented with 10% of fetal bovine serum (FBS), streptomycin (50 lg/ mL), penicillin (50 U/mL), 1% non-essential amino acids and 1 mM Sodium Pyruvate, and incubated at 37 °C in humidified atmosphere containing 5% CO2. The medium was replaced every 2 or 3 days. When the cultures were about 85–90% confluent, the cells were trypsinized by 0.05% trypsin and 0.53 mM EDTA solution and incubated for 5 min at 37 °C in humidified atmosphere containing 5% CO2. Trypsinization was stopped by adding 20% FBS, resuspended and plated in flasks fed with fresh basic complete media. Cells were seeded again at 1:4 density ratio and incubated at 37 °C in humidified atmosphere containing 5% CO2. 2.10.2. Treatment of the cultures A lot of U87MG cancer cell lines were placed at the final density of 60 104 cells/well of a 96-multiwell flat-bottomed 200-ll microplates and untreated or treated with different concentrations of free drugs, unloaded and drug-loaded NLC. We also treated some cultures with an equal volume of DMSO used for the solubilization of IDE and FA (35 lM). 2.10.3. MTT bioassay MTT bioassay was performed to monitor cell viability (Martins et al., 2012; Carbone et al., 2012b; Musumeci et al., 2006). A lot of cell lines cultures were exposed to free drugs, unloaded and FA-loaded (0.7% w/w) NLC-20% (20, 36, 40 lM) for 24, 48 or 72 h [4], in order to establish their optimal concentrations and the exposure times (Campisi et al., 2003). The study was accomplished by investigating cell death effects of the same tested concentrations of free IDE and IDE-loaded NLC, used for comparative purposes (Tai et al., 2011). At the end of treatment time, 20 ll of 0.5% of MTT salts in PBS were added to each well sample. After 1 h of incubation, the supernatant was removed and replaced with 100 ll dimethylsulfoxide and incubated at 37 °C in a humidified 5% CO2–95% air mixture for 1 h. The optical density of each well sample was measured with a microplate spectrophotometer reader (Titertek Multiskan; Flow Laboratories, Helsinki, Finland) at k = 570 nm. Three measurements for each samples were performed. Results were expressed as a percentage of the control (untreated cells), taken as 100%, to equalize the different values obtained. 2.10.4. Immunocytochemistry A lot of U87MG cancer cell lines were placed in Lab-Tek II Chamber-Slide Systems (Nalge Nunc International, Germany) at the final density of 0.5 105 and incubated at 37 °C in a humidified 5% CO2–95% air mixture. When the cell cultures arrived about at 80% of confluence, they were untreated or treated with the optimal concentration of free drugs (FA or IDE), unloaded and drug-loaded NLC and incubated at 37 °C in a humidified 5% CO2–95% air mixture for 24 h. After the treatment, the cell cultures were fixed for 20 min with 4% paraformaldehyde in 0.1 M PBS, they were treated with 1% NGS in PBS and incubated for 1 h at 37 °C in humidified air and 5% CO2, in order to block unspecific sites (Campisi et al., 2004). Then, they were washed three times with PBS, were successively added with mouse monoclonal antibody against caspase-3 (1:100) and incubated overnight at 37 °C in humidified air and 5% CO2. Cell cultures were then washed three times with PBS and incubated for 2 h with TRITC-conjugated anti-mouse IgG polyclonal antibody (1:64
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
in PBS). Finally, the cell cultures were washed three times with PBS, and the Lab-Tek II Chamber-Slide Systems were mounted in PBS/glycerol (50:50), and analyzed on a Leica fluorescent microscopy (Germany), in order to visualize caspase-3 positive cells. No non-specific staining of U87MG cancer cell lines was observed in control incubations in which the primary antibody was omitted (Campisi et al., 2003; Campisi et al., 2004; Campisi et al., 2012). 2.11. Data analysis Results of the technological characterization were expressed as the mean ± S.D. of three experiments and Student’s t-test was used to evaluate the significance of the difference between mean values. Biological data were statistically analyzed using one-way analysis of variance (one-way ANOVA) followed by post hoc Holm–Sidak test to estimate significant differences among groups. Data were reported as mean ± SD of four experiments in duplicate. In both statistical analysis, differences between groups were considered to be statistically significant for p < 0.05 and highly significant for p < 0.01. 3. Results and discussion A preliminary physico-chemical characterization was performed to identify the proper raw materials, among those selected on the basis of our previous studies on NE and SLN, for the preparation of NLC with small particle size and high homogeneity. 3.1. Preliminary screening on raw materials for NLC preparation To establish the feasibility to obtain NLC, a preliminary screening was performed on different surfactants and oils selected on the basis of previous experiments (Montenegro et al., 2006; Carbone et al., 2012b). As solid lipid matrix we chose CP due to its high safety for the systemic administration and its ability to obtain SLN (Lukowski et al., 2000; Yang et al., 2011). All NLC were prepared by a low energy organic solvent-free phase inversion process (PIT method). PIT values ranged from 65 to 73 °C for all the prepared NLC and were not significantly affected by the primary surfactant. Nevertheless, a slight increase of PIT temperature was observed with the increase of the oil lipophilicity (data not shown), thus suggesting that different interactions occurred between the liquid lipid and the other components. The effects of the selected variables were explored on both particle size and polidispersity by AFM and PCS analysis. As shown by the physico-chemical characterization (Fig. 2), the surfactant significantly affected the mean particle size, since all samples prepared with isoceteth-20 had a PDI > 0.3%, related with the presence of two or three peaks in size distribution. AFM images confirmed PCS results, revealing the presence of a heterogeneous system (Fig. 3A and B).
15
Both ceteth-20 and oleth-20 allowed the formation of NLC with a single peak in size distribution (PDI < 0.3%), even if oleth-20 generates particles with mean size smaller than those obtained with ceteth-20 (Fig. 2). A similar behavior was reported for SLN and was attributed to the presence of the unsaturated bond in both the chemical structure of surfactant and cosurfactant, which reduced the curvature radius at the interfaces favoring the formation of small particles (Montenegro et al., 2006). Student’s t-test was performed with a 99% level of confidence, to test whether the differences between ceteth-20 and oleth-20 were statistically significant. The difference between ZAve values were found to be highly significant (p < 0.01), thus encouraging the use of oleth-20, instead of ceteth-20, in the preparation of NLC. Instead of primary surfactants, it seems that the oil lipophilicity did not significantly affect particle size distribution, even if the use of IPS allowed to obtain the smallest particle size ( NE > SLN > NLC-20% (Fig. 8). It is possible that the addition of a little amount of oil could interact with the wax so as to reduce the path’s length that the drug must make for diffusion through the lipid matrix, thus determining greater amount of FA released after 24 h (Thatipamula et al., 2011). On the contrary, increasing the amount of oil content (NLC-20%), allowed to achieve a better control in drug release, as supported by DSC data. The release experiment performed on NLC loaded with FA 1.2% w/w, confirmed these data.
18
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
Fig. 7. Thermograms of FA-loaded freeze-dried SLN, NLC 10% and NLC 20%.
(Van Meir et al., 2010) (Table 2). Previously, unloaded NLC (with 20% of oil content) was studied at different concentrations (20, 36, 40 lM) to evaluate its effect on cell viability. The study was accomplished by investigating cell death effects of the same tested concentrations of free IDE and IDE-loaded NLC, used for comparative purposes (Tai et al., 2011). No significant difference between untreated and DMSO-treated culture cells was found. Unloaded NLC showed no cytotoxic effects on U87MG cell cultures at all the tested concentrations. Results of the observations through fluorescent microscope and MTT assay highlight significant changes in cell viability related to the tested concentrations and further differences between free drugs and drug-loaded NLC (Table 2). The optimal values of concentration and exposure time for the treatment were found to be respectively 36 lM and 24 h, which are in agreement with other findings (Tai et al., 2011; Karthikeyan et al., 2011). Data showed that at the selected concentration, FA was able to reduce cell viability much more than IDE, whose cytotoxic effect was previously reported on dopaminergic neuroblastoma cells (Tai et al., 2011). The increase of the cytotoxic effect observed when drugs were loaded
Table 2 MTT assay on U87MG human cell untreated or treated with DMSO or different concentrations (20, 36, 40 lM) of surfactant solution, free drugs (IDE and FA), unloaded and drug loaded NLC (containing 20% of oil in the lipid matrix). Results are expressed as a percentage of the control (untreated cells), taken as 100%, to equalize for different values between the treatments. Results are reported with standard deviation (±S.D.). Fig. 8. Release profile from FA-loaded NLDDS in comparison: NE, SLN and NLC with different oil content containing 0.7% w/w of FA (A); NLC with 10% and 20% of oil content containing FA 1.2% w/w (B).
Treatment
Control DMSO Surfactant solution IDE FA NLC IDE-NLC FA-NLC
Due to its small particle size and better control in drug release, NLC-20% was chosen for in vitro further investigation, in order to evaluate its efficacy on human glioblastoma cells (U87MG). 3.5. In vitro biological test To assess if free FA or FA-loaded NLC were able to induce cytotoxicity, MTT bioassay was performed on U87MG culture cell lines
* **
% Vitality cells ± S.D. 20 lM
36 lM
40 lM
100 99.97 ± 0.02** 99.95 ± 0.01** 91.19 ± 0.03* 99.94 ± 0.06* 99.07 ± 0.03* 89.96 ± 0.04* 79.43 ± 0.05*
100 99.76 ± 0.01** 99.98 ± 0.01** 68.89 ± 0.02* 69.81 ± 0.04** 99.04 ± 0.01** 51.25 ± 0.04** 38.96 ± 0.02**
100 99.81 ± 0.03** 99.91 ± 0.03** 87.89 ± 0.01* 84.58 ± 0.04** 71.04 ± 0.01** 72.86 ± 0.01** 64.64 ± 0.01*
Significant for p < 0.05. Highly significant for p < 0.01.
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
19
Fig. 9. A: Fluorescent microscopic analysis of caspase-3 cleavage in U87MG human cell lines in absence or in presence of DMSO (35 lM) or 36 lM of free IDE, free FA, unloaded NLC, IDE-loaded NLC or FA-loaded NLC, for 24 h; B: quantification and statistical analysis of caspase-3 positive U87MG human cell lines obtained and collected from 4 fields/coverslip of four separate experiments. Highly significant values for p < 0.01**.
into the selected NLC highlights the potentiality of the nanoparticulate system in increasing drug effectiveness. To verify if the drug alone or loaded into the NLC was able to activate the apoptotic pathway, we assessed caspase-3 cleavage by immunocytochemical analysis (Fig. 9A). No significant number of positive cells for caspase-3 in DMSO and unloaded NLC was observed (Fig. 9). In contrast, we observed a significant number of positive cells for caspase-3 in the cultures treated with FA or IDE as free drugs (Fig. 9A). The number of positive cells for caspase-3 significantly increased when the drugs were loaded into the NLC (Fig. 9), and resulted in a major induction of cell death in U87MG cancer cell lines. The effect of free FA or FA-loaded NLC appeared more evident when compared to free IDE or IDE-loaded NLC, respectively (Fig. 9). The quantification and statistical analysis of caspase-3 immunolabeling obtained and collected from 4 fields/coverslip of four separate experiments was reported in Fig. 9B. No non-specific staining of U87MG cell line cultures was observed in control incubations in which the primary antibody was omitted. Taken together these data demonstrate that FA is able to induce cytotoxic effects in human glioblastoma cancer cell lines activating the apoptotic pathway, and that the effects significantly increase in FA-loaded NLC. 4. Conclusion The low energy organic solvent-free phase inversion process used in this study allowed to obtain a NLC system with small particle size and high homogeneity when oleth-20 was used as
primary surfactant. DSC studies on SLN and NLC containing 10% or 20% of oil component in respect to the total lipid mixture, highlight the formation of a different lipid matrix related not only to the oil presence or not (NLC vs SLN) but also to the amount of the oil used for the preparation of the NLC. In vitro release studies showed a better control in FA release rate for NLC with the higher percentage of oil, which was comparable to those observed for SLN. MTT bioassay and caspase-3 cleavage performed on human glioblastoma cancer U87MG cells, showed no cytotoxic effect of the unloaded tested NLC and the activation of the apoptotic pathway caused by FA-loaded NLC, whose effectiveness was higher compared to the free drug. Furthermore, a greater effect was observed when cells were treated with FA or FA-loaded NLC, compared to IDE or IDE-loaded NLC. These results strongly support that FA-loaded NLC could be potentially used for the treatment of glioblastoma. Further studies are ongoing to better understand the molecular mechanism induced by free FA or FA-loaded-NLC. Acknowledgments The present work was financially supported by Lilly Foundation and Italian MIUR: PRIN 2010-2011 for University of Catania. We thanks BASF Italia S.p.A. for purchasing cetyl palmitate. References Anton, N. et al., 2007. Nano-emulsions and nanocapsules by the PIT method: an investigation on the role of the temperature cycling on the emulsion phase inversion. Int. J. Pharm. 344, 44–52.
20
C. Carbone et al. / European Journal of Pharmaceutical Sciences 52 (2014) 12–20
Bandugula, V.R., Rajendra Prasad, N., 2013. 2-Deoxy-d-glucose and ferulic acid modulates radiation response signaling in non-small cell lung cancer cells. Tumor Biol. 34, 251–259. Campisi, A. et al., 2003. Glutamate-induced increases in transglutaminase activity in primary cultures of astroglial cells. Brain Res. 978, 24–30. Campisi, A. et al., 2004. Glutamate-evoked redox state alterations are involved in tissue transglutaminase upregulation in primary astrocyte cultures. FEBS Lett. 578, 80–84. Campisi, A. et al., 2012. Expression of tissue transglutaminase on primary olfactory ensheathing cells cultures exposed to stress conditions. Neurosci. Res. 72, 289– 295. Carbone, C. et al., 2012a. Chemical and technological delivery systems for idebenone: a review of literature production. Expert Opin. Drug Deliv. 9, 1377–1392. Carbone, C. et al., 2012b. Preparation and optimization of PIT solid lipid nanoparticles via statistical factorial design. Eur. J. Med. Chem. 49, 110–117. Karthikeyan, S. et al., 2011. Radiosensitizing effect of ferulic acid on human cervical carcinoma cells in vitro Toxicol. In Vitro 25, 1366–1375. Kaur, I.P. et al., 2008. Potential of solid lipid nanoparticles in brain targeting. J. Control. Release 127, 97–109. Kovacevic, S. et al., 2011. Polyhydroxysurfactants for the formulation of lipid nanoparticles (SLN and NLC): effects on size, physical stability and particle matrix structure. Int. J. Pharm. 406, 163–172. Kundu, P., Mohanty, C., Sahoo, S.K., 2012. Antiglioma activity of curcumin-loaded lipid nanoparticles and its anhanced bioavailability in brain tissue for effective glioblastoma therapy. Acta Biomat. 8, 2670–2687. Lukowski, G. et al., 2000. Crystallographic investigation of cetylpalmitate solid lipid nanoparticles. Int. J. Pharm. 196, 201–205. Martins, S. et al., 2012. Brain delivery of camptothecin by means of solid lipid nanoparticles: formulation design, in vitro and in vivo studies. Int. J. Pharm. 439, 49–62. Montenegro, L. et al., 2006. Effect of oil phase lipophilicity on in vitro drug release from O/W microemulsions with low surfactant content Drug Dev. Ind. Pharm. 32, 539–548.
Montenegro, L. et al., 2011. In vitro evaluation of idebenone-loaded solid lipid nanoparticles for drug delivery to the brain. Drug Dev. Ind. Pharm. 37, 737–746. Müller, R.H., Radtke, M., Wissing, S.A., 2002. Nanostructured lipid matrices for improved microencapsulation of drugs. Int. J. Pharm. 242, 121–128. Müller, R.H., Shegokar, R., Keck, C.M., 2011. 20 Years of lipid nanoparticles (SLN & NLC): present state of development and industrial application. Curr. Drug Discov. Technol. 8, 207–227. Musumeci, T., Vicari, L., Ventura, C.A., Gulisano, M., Pignatello, R., Puglisi, G., 2006. Lyoprotected nanosphere formulations for paclitaxel controlled delivery. J. Nanosci. Nanotechnol. 6, 3118–3125. Preetz, C. et al., 2010. Application of Atomic Force Microscopy and ultrasonic resonator technology on nanoscale: distinction of nanoemulsions from nanocapsules. Eur. J. Pharm. Sci. 39, 141–151. Serafim, T.L. et al., 2011. Oliveira, Lipophilic caffeic and ferulic acid derivatives presenting cytotoxicity against human breast cancer cells. Chem. Res. Toxicol. 24, 763–774. Ò Ò Severino, P. et al., 2012. Crystallinity of Dynasan 114 and Dynasan 118 matrices for the production of stable Miglyol-loaded nanoparticles. J. Therm. Anal. Calorim. 108, 101–108. Shah, V.P. et al., 1989. Determination on in vitro drug release from hydrocortisone creams. Int. J. Pharm. 53, 53–59. Stancampiano, A.H.S. et al., 2006. Technological and biological characterization of idebenone-loaded solid lipid nanoparticles prepared by a modified solvent injection technique. J. Biomed. Nanotech. 2, 53–260. Tai, K.K., Pham, L., Truong, D.D., 2011. Idebenone apoptotic cell death in the human dopaminergic neuroblastoma SHSY-5Y cells. Neurotox. Res. 20, 321–328. Thatipamula, R.P. et al., 2011. Formulation and in vitro characterization of domperidone loaded solid lipid nanoparticles and nanostructured lipid carriers. DARU 19, 23–32. Van Meir, E.G. et al., 2010. Exciting new advantages in neuro-oncology: the avenue to a cure for malignant glioma. CA Cancer J. Clin. 60, 166–193. Yang, R. et al., 2011. The influence of lipid characteristics on the formation, in vitro release, and in vivo absorption of protein-loaded SLN prepared by the double emulsion process. Drug Dev. Ind. Pharm. 37, 139–148.
